Welcome to the first day of STT2860 Intro to Data Management and Visualization with Dr. Jill Thomley. To prepare for the start of our exciting journey…

• Log on to AsULearn!

• Find and open the link “Course Syllabus”

• Find and open the link “Day 01 Introduction Slides”

• Begin to familiarize yourself with the course documents and our AsULearn site. AsULearn is the vital hub of our course for assignments and communication.

We will do some activities to make sure we are set up on course technology and familiar with course structure / expectations.

• Medical research in high school at Medical College of Ohio
• BA in Psychology from Harvard — focus in social psychology
• MS in Industrial / Organizational Psychology from RPI
• Health care research at the VA hospital in Albany, New York
• PhD in Decision Sciences and Engineering Systems from RPI
  • statistics and operations research
  • information systems and databases
• Statistical consulting in a variety of disciplines
• Grant evaluation, primarily in STEM education
• Basic, Excel, Fathom, StatCrunch, SPSS, SAS, Minitab, R, Git…
• Also, I just completely ♥ data!

* not true, I am kind of a nerd, sorry-not-sorry

Though truthfully, a better name would be Miss Fatty McCatty :)

From: ASA Statement on the Role of Statistics in Data Science

The foundations of data science include, but are not limited to: database management, statistics/machine learning, distributed and parallel computing. /multidisciplinary/

“Framing questions statistically allows us to leverage data resources to extract knowledge and obtain better answers.” /finding answers and solving problems/

“Statistical methods aim to focus attention on findings that can be reproduced by other researchers…” /reproducible research/

“…next generation of statistical professionals needs a broader skill set and must be more able to engage with database and distributed systems experts.” /beyond point-and-click/

“Starting September 1 [2016], JASA ACS will require code and data as a minimum standard for reproducibility of statistical scientific research. New infrastructure is being established to support this initiative … A new editorial role—associate editor for reproducibility (AER)—will be added to ensure we meet a standard of reproducibility.” (Reproducible Research in JASA)

ASA also released Recommendations to Funding Agencies for Supporting Reproducible Research in January 2017, suggesting more statistician reviewers to assess quality of plans for data management and research reproducibility.

Potential barrier: researchers may lack programming and “best practices” skillsets. Also, some disciplines or organizations may not recognize computer data and code as research products in and of themselves.

The Chapel was acquired by RPI in 1958 and transformed into the VCC in 1976. President Low stated it was “the most unique computing center in the world.” (There was a definitely a lot of invoking of deities, in my experience…)

Where are you in this diagram? Let’s explore this for ourselves.

Note: This table was created and formatted using RMarkdown. All of the slides were. They are not MS PowerPoint. These skills are among those you will learn in this class.

My PhD advisor once said, “We can make a statistician out of anyone”. I think that’s even more true of data science! I have written several letters of recommendations for students from this class to graduate programs such as NC State.

R : a free and open-source statistical programming language

R Studio : a free and open-source IDE for R, it supports projects and add-ins and has integrated support for Git and GitHub

tidyverse : “an opinionated collection of R packages designed for data science”

Use your ASU Username and Password to log in to the RStudio server. Let me know if you have any trouble! Also make sure to complete “Confirm ASU RStudio server login” on AsULearn.

DataCamp is an “intuitive learning platform” for data science. Learn concepts and R code via short videos and hands-on-the-keyboard exercises, with feedback on each exercise.

Multiple tracks for people with different interests and needs.

ASA on Data Science: “Statistical education and training must continue to evolve…”

We will use DataCamp extensively for learning and practice. It’s how I initially learned many of my data science skills. We’ll also explore some of the tracks related to Data Science.

• Sign in or create an account on DataCamp. Use your ASU email address.

• Go into your profile and add your first and last name so it shows up on the class roster.

• Join our DataCamp class. The link for this is on AsULearn.

• Finally, complete “Confirm DataCamp account creation” on AsULearn sometime after class is over.

You will have six months of free access to DataCamp, during which you can access all content — not just courses I assign.

Git : an open-source version control and collaboration system

GitHub : web-based hosting service (hub) for version-controlled projects using Git, with some added features

You will learn more about this resource in upcoming weeks.

Except for extreme emergencies, all written communication must be handled through your Private Forum on AsULearn. This will be new for most of you.

I prefer this method because it stores all exchanges in a place where we can easily access them all semester long and it keeps my class communications separate from the rest of my email (which can turn into a black hole).

Let’s go to AsULearn and take a look!

The R for Data Science digital textbook “will teach you how to do data science with R: You’ll learn how to get your data into R, get it into the most useful structure, transform it, visualise it and model it…Just as a chemist learns how to clean test tubes and stock a lab, you’ll learn how to clean data and draw plots—and many other things besides.”

The R Programming for Data Science digital textbook “is about the fundamentals of R programming. You will get started with the basics of the language, learn how to manipulate datasets, how to write functions, and how to debug and optimize code. With the fundamentals provided in this book, you will have a solid foundation on which to build your data science toolbox.”

The Modern Dive digital textbook “assumes no prerequisites: no algebra, no calculus, and no prior programming/coding experience. This is intended to be a gentle introduction to the practice of analyzing data and answering questions using data the way data scientists, statisticians, data journalists, and other researchers would.”

Shortly we will switch over to the course syllabus and talk some more about objectives and assessments.

As we review, remember that STT2860 is a 3 credit hour course. The standard expectation for a college course is that you spend 2 to 3 hours outside of class for every 1 hour spent in class.

3 + ((2 to 3) × 3) = 9 to 12 hours per week

Weeks where you have a full DataCamp course, that will take up about four hours of time. In-class time is only 2.5 hours/week.

Basic mathematics proficiency is the only course prerequisite. You are not expected to have stats or programming experience. If you have one or both, you might not require as much time to master course concepts, but do not assume until you try.

Explorations

• engage with digital textbooks and other online materials
• watch assigned videos and participate in other activities
• come to class, take organized notes, actively participate
• use office hours to get individualized attention/feedback

Assessments

• complete assigned DataCamp modules
• complete assigned homework and projects
• practice new and existing data science and R skills
• solidify knowledge and make new connections
• review and understand your misconceptions
• demonstrate content and skill proficiency

One of our goals in this class is reproducible research and code; this includes learning to make our code readable to others.

Coding is often multi-stage: we get something that works first, then we refine it to make it more efficient and readable.